Abstract

The cloud computing paradigm enables the work anywhere any time paradigm by allowing application execution and data storage on remote servers. This is especially useful for mobile computing and communication devices that are constrained in terms of computation power and storage. Relying more on powerful remote servers further allows each individual device's hardware to have reduced functionality, thus making them cheaper. Though there could be energy and performance benefits for mobile devices by offloading application tasks to remote servers, it is not clear whether there would be reduction in total end-to-end energy consumed when including energy consumed at the remote server and the network in between. The goal of this paper is to characterize the overall end-to-end energy consumed for task offloading in mobile devices. An analytical model of end-t o-end energy consumed for task offloading is proposed and evaluated to understand the theoretical limits within which task offloading can save energy. Additionally, an empirical measurement-based evaluation is done with common off-the- shelf mobile devices to determine whether task offloading is more energy-efficient end-to- end compared to a scenario where the task is executed locally. Introduction. Computing with mobile devices have always presented challenges in terms of storage, memory, processing, network connectivity, bandwidth, and battery lifetime in comparison to their static counterparts like desktop computers. With the technological advances in recent years improving ubiquitous connectivity and bandwidth, cloud computing has become feasible allowing these constrained devices to utilize the greater storage, memory, and processing capabilities of powerful remote servers.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call